When money is being taken from a parent

What it actually looks like

The picture most families carry is a scam phone call. Those exist, and they are the minority. The larger share involves a family member, a new friend, a paid caregiver or a neighbor who has become indispensable — and in most cases the older adult is handing the money over, not having it stolen behind their back.

Signals that are worth acting on

• New account activity that does not match a lifetime of habits: cash withdrawals from someone who never used cash, wire transfers, gift cards, cryptocurrency.
• A new person who has become the one who drives to appointments, handles the post, and is present for every phone call.
• Documents that have changed recently — a new power of attorney, a new name on a deed or an account, a changed beneficiary — especially where the change followed an illness.
• Bills going unpaid while money is clearly moving.
• Isolation. A previously sociable parent who now sees almost no one, where the new person explains why each visit is a bad time.
• Secrecy and defensiveness about money that was never a private subject before.

Isolation is the signal that most reliably precedes the rest. If someone is managing who gets to speak to your parent, treat that as the thing to address first, before the money.

Seven situations that look identical from outside

Money leaving an older person's account at a rate that alarms their children is the symptom, not the diagnosis. These seven have different evidence, different levers and different people to call, and the single most useful thing a family can do early is work out which one they are looking at — including the possibility that it is the first.

1. Spending you disagree with. An adult with capacity may spend, lend and give away their own money, and being wrong about it is their right. There is no lever here, and looking for one damages the relationship you will need if something else is happening.
2. A scam by a stranger. Someone impersonating a grandchild, a government agency, a suitor or a contractor. The parent usually authorizes the payment, which is what makes it hard; the reporting path runs to law enforcement and the fraud reporting systems, and the immediate work is stopping what has not yet moved.
3. Transactions nobody authorized. A card used without permission, a forged check, an account opened in the parent's name. This is the narrowest category and the one most likely to be reversed, because it is the situation consumer fraud processes were built around. Speed decides the outcome.
4. An agent misusing a power of attorney. The transactions are authorized on their face because the document authorizes them. The question is whether the authority is being exercised for the parent or for the agent, and it has its own section below.
5. A caregiver or family member with access. Access granted informally — a card handed over for shopping, a name added to an account for convenience — and then used beyond what was intended. The evidence is the pattern of use against what was agreed, and it is rarely documented.
6. Undue influence. Someone has made themselves the channel through which the parent sees the world, and the parent's decisions have become that person's decisions. The transfers may be perfectly voluntary in form. What is documented here is the isolation, not the payment.
7. Judgment failing with cognitive decline. No one is exploiting anyone; the parent is making decisions they would not have made five years ago, repeatedly and to their own cost. This calls for a clinical conversation and a structural fix, not a report.

These overlap, and more than one is often true at once — a scammer finds a parent whose judgment has already slipped, and a relative steps in to help and then stays too long. Treat the classification as a working hypothesis you keep revising, not a verdict.

A family disagreeing about money is not evidence that anyone is stealing. Siblings routinely reach opposite conclusions about the same set of transactions, and an accusation made from suspicion alone is difficult to withdraw. What distinguishes exploitation from a bad decision is not how much money moved, but whether the decision was actually the parent's.

Why it is hard to stop, even once you see it

An adult with capacity is entitled to give their money away, to anyone, for any reason, including a reason their children think is foolish. That is not a loophole — it is the same right that lets them buy a boat or fund a grandchild's tuition. The legal question is rarely "is this a bad decision" and almost always "is this decision actually theirs".

That distinction is why the strongest evidence is usually not the transfer itself but the pattern around it: undue influence, isolation from other advisers, a change made days after a hospital stay, or a person who cannot explain their own transaction.

Accusing the wrong person, or the right person too early, usually ends with the older adult defending them and closing the door on you. The people who intervene successfully almost always secure the accounts and gather the record before they open the confrontation.

The brake nothing offers you

Reporting tells an agency. It does not stop the next withdrawal, and the gap between those two things is where the money usually goes. A brokerage holding the account can stop it.^{Source 2}

FINRA's rule lets a firm place a temporary hold on a disbursement or a securities transaction where it reasonably believes financial exploitation has occurred, is occurring, or will be attempted. The standard is the firm's reasonable belief, not proof — which means a family does not have to have established anything before asking the firm to look.^{Source 2}

The rule's own definition of exploitation reaches conduct "through the use of a power of attorney, guardianship, or any other authority". That is a frequent shape of the problem: the person moving the money has the legal right to move it. Having a valid power of attorney is not an answer to the question, and the rule says so.^{Source 2}

The firm must notify people within two business days — but it may skip anyone it reasonably believes is involved in the exploitation, including a party authorized on the account. So a firm is not obliged to warn the suspected exploiter that it has frozen them, which is the difference between a useful hold and a warning shot.^{Source 3}

The form to fill in before anything happens

Notification also runs to the Trusted Contact Person — a name the account holder gives the firm in advance. If nobody was named, there may be no one outside the household for the firm to call. Adding a trusted contact takes a phone call, gives that person no authority over the account and no right to trade it, and is the single cheapest protective step available on the planning side.^{Source 3}

What to ask an ordinary bank, and why they can say yes

A checking account is not covered by the brokerage rule, and families often read a bank's caution as unwillingness. The more useful frame is that federal law has already removed the risk of reporting. A trained employee in a supervisory, compliance or legal role — the statute names the Bank Secrecy Act officer specifically — is protected from civil and administrative liability for disclosing suspected exploitation of an older customer to a covered agency, provided they act in good faith and with reasonable care, and the institution is protected behind them where that training was given.^{Source 15}

So ask for the compliance officer or the Bank Secrecy Act officer by name rather than arguing at a branch counter. Those are the roles the immunity is written around, and they are the people for whom reporting is a protected act rather than a judgment call.^{Source 15}

Do not mistake this for a brake. The federal protection covers disclosure — it creates no power to hold, delay or reverse a transaction, and the statute says it does not reach anything that is not a disclosure. Sahvelo has not established that any state it has read gives an ordinary bank a hold power equivalent to the brokerage rule. Ask the bank what its own policy allows, and treat a hold on a bank account as something it may choose to do rather than something you can require.^{Source 15}

Which rule is doing the work

The levers on this page come from different places, and they do not cover the same accounts. Asking a bank for something that only exists in the brokerage rulebook is how a family concludes that nothing can be done.

• The FINRA rule is a broker-dealer rule. It reaches investment accounts at a member firm — a brokerage, and the brokerage arm of a bank. It does not reach an ordinary checking or savings account.
• State power-of-attorney statutes reach both banks and brokerages, and they are where the answer lives when the person moving the money is the agent. They vary more than a single federal-sounding rule would suggest; the table further down gives the states Sahvelo has read.
• State mandated-reporter law decides who is obliged to report, and in California it makes the bank employee you are speaking to a reporter in their own right.
• Everything else is the institution's own policy. Alerts, a second signatory, a card reissued, a note on the file: none of these is compelled by a rule Sahvelo can cite, and all of them are routinely available for the asking. Policy is not weaker than law here — it is faster.

Ask which of these the institution is relying on, and get the answer in writing where the rule provides for writing. A refusal recorded as “bank policy” and a refusal recorded under a statute are not the same thing later, and only one of them has a remedy attached.

When the person taking the money is the one holding the power of attorney

This is a frequent version of the problem and the one families are least prepared for, because the document they were told to be relieved about is the instrument being used. It helps to see that it raises two separate questions, with different answers and different evidence.^{Source 2}

1. Was the document itself obtained properly? A power of attorney signed by someone who could not understand what they were signing, or who was pressured into it, is vulnerable to challenge on that ground alone.
2. Is a validly held authority being exercised wrongly? Here the document is sound and the conduct is not, and the question is whether the agent is converting the money or acting outside what was authorized.

The two paths call for different things. The first is a challenge to the instrument, and it turns on what the parent understood on the day they signed. The second does not require unwinding the document at all: an agent owes duties to the person who appointed them, and can be made to account for what they did with the authority whether or not the appointment was valid.

If your parent still has capacity, revoking the power of attorney is the fastest protective step available and does not require anyone to agree about what happened. It is also the step most often delayed, because it feels like an accusation. It is not one — a principal may revoke for any reason or none.

What to ask the institution to do

An institution presented with an agent it has reason to distrust is not obliged to keep honoring the document, and in some of the states Sahvelo has read, a report of exploitation by the agent is written into the statute as a ground for refusing it. That is the practical link between reporting and stopping the money: the report is not only a referral to an agency, it can also be the fact that lets the bank say no.^{Source 9Source 11}

• Ask whether the institution will decline to act on the power of attorney, and on what stated ground.
• Tell it plainly if a report has been made, and to which agency — in Florida and New York that fact is itself the ground the statute names.
• Ask for the refusal in writing where the state requires a written reason, and keep it.
• Ask what happens to transactions already in flight, which is a different question from what happens next.

This cuts both ways, and it is why the page does not tell you to report as a tactic. A refusal wrongly given exposes the institution — Florida and New York both allow a court to order acceptance and to shift legal costs onto the party that refused. The same machinery protects an agent who is doing nothing wrong, which is the correct outcome when a family is mistaken about who is at fault.^{Source 10Source 12}

There is an instrument working against you here, and a family should know it exists before they are surprised by it. In Arizona and California an agent may sign an affidavit stating they had no actual knowledge that the power was revoked or the principal had died. Arizona treats that as a rebuttable presumption the power is still live. California treats it as conclusive proof, as to acts undertaken in good faith reliance on it. It is a legitimate tool — it is how an honest agent gets a hesitant bank to act — and it is equally available to an agent who should not be acting at all, because it removes the institution's residual risk in honoring the document. Where you are asking an institution to stop honoring a power of attorney, expect the affidavit to be the answer, and put the reason for distrust to them in writing so their reliance is no longer unqualified.^{Source: A.R.S. §14-5505 — the agent's affidavit of non-revocation (opens in a new tab)•Source: Cal. Prob. Code §4305 — the attorney-in-fact's affidavit as conclusive proof (opens in a new tab)•}

What your state gives you here varies more than almost anything else on this page — Arizona moves a burden of proof, California routes the question through the institution's own good faith, Florida and New York name the report in the statute. The table below sets out the four Sahvelo has read.

What to do, in order

1. Address the isolation before the money. Whoever is controlling access is the thing to work on first, and confronting them about finances usually tightens that control.
2. Write down what you have actually seen, with dates — transfers, new names on accounts, a document signed shortly after a hospital stay, a caregiver who now drives to every appointment.
3. Call the financial institution and ask what it can do. For a brokerage, that includes a temporary hold and an internal review; ask for both by name.
4. Ask whether a trusted contact is on file, and if your parent is willing, add one now.
5. Report to adult protective services or the hotline your state uses. Reporting is not accusation and in most states you need suspicion rather than evidence.
6. Preserve the paper: statements, the power of attorney, any new will or deed, texts and emails. Recovery cases turn on the pattern, not on a single transfer.
7. Get a lawyer involved where a document has been changed, where a power of attorney is being used, or where property has moved. Those are the situations where waiting makes recovery harder.

Do not confront the suspected person first. The people who intervene successfully almost always secure the practical protections — the hold, the trusted contact, the report — before anyone is accused, because an accusation that lands badly ends with the older adult defending them and closing the door.